Estimates based on field data have indicated representative hydrofracture toughness parameters of one to two orders of magnitude higher than the values determined from conventional laboratory tests . Several studies on quarried rocks show a significant increase in mode I fracture toughness with an increase in the confining pressure [2-3]. The measured data show considerable scatter, but an increase that is roughly linear with confining pressure is observed . These tests, however, were not carried out for a fracture generation in a hydraulic fracture environment. The increase of fracture toughness with confining pressure is generally attributed to a change in the behavior of the process zone which exists ahead of the stress free crack tip. Earlier studies at CESE have documented strong correlations between elevations in K?c and fluid rheology. However, these measurements did not provide an adequate separation of direct confining-pressure effects from indirect effects due to fluid flow and leak-off behavior. In the current study apparent hydrofracture toughness parameters were determined from measurements carried out in a hydraulic fracture environment, but for which complicating factors from fluid-flow effects were minimized. An initial set of measurements was carried out with different fracture fluids of different rbeologics, to dcfmc a set of appropriate test conditions for the pressure-dependence determination. The suite of measurements reported in this paper details the effects of confining pressure on behavior observed during fracture initiation and fracture propagation, in both wax-sealed and kerosene-saturated specimens.
Techniques were developed and implemented for fabricating test specimens with a range in initial cavity diameters, between the normal 1.5" size used for most of the earlier tests, up to a four-inch diameter, in order to assess the importance of potential scale-dependent factors such as the relative sizes of the fracture surface and a process zone at the crack tip. The overall objective of the program was to develop reliable procedures for estimating hydraulic fracture performance in the field.
A schematic drawing of the test specimen configuration is presented in Figure 1. Cylindrical test specimens 7« inches in diameter and 12 inches in length were prepared with a central borehole halfway through the specimen, and a horizontal cavity at the base of the borehole. Test specimens were prepared from blocks of Sciota Sandstone and Indiana Limestone specimens. The details of high-pressure hydrofracture test system used for this study has been described elsewhere . The test sequences began with the application of confining pressure and an approximately two-hour equilibration period. The pore fluid system was vented to atmospheric pressure during equilibration and testing. The fracture fluid was loaded into the pressure intensifier, and the low-pressure side of the intensifier was pressurized by a kerosene-f'dled pump. Flow rates were determined by direct volumetric measurements. The tests were continued until evidence for stable fracture growth or fracture break-through was observed.
Post-test visual observations of the fracture surface were carried out after breaking open the specimen along the hydrofracture plane. Three fracture fluids were used. All fracture fluids were prepared immediately prior to the tests. Three of the scoping tests and all subsequent pressure-dependence tests were carried out with the 50-pound linear gel mixture described in Holder and Gray . Three-inch and four-inch diamond cutting wheels were used to cut large-diameter cavities in some of the test specimens, in order to determine the effects of cavity